Instability of hydrogen peroxide solutions was noted during the experimental exposure of human cells in culture to hydrogen peroxide in experiments designed to study the production and repair of DNA single-strand breaks. A hydrogen peroxide concentrate was diluted into culture medium, which was then added to experimental cell cultures at various times, with all cultures assessed for DNA damage at 2 h. Only cells treated by the first addition had observable DNA damage. This result was unexpected since these cells had had the maximum repair time. It was determined that the hydrogen peroxide had been eliminated by the culture medium. To determine the mechanism of this elimination, 200 microM hydrogen peroxide was added to various cell culture components, and the solutions were assayed for hydrogen peroxide after 1 h at 37 degrees C. Although most components (except the balanced salts) showed some hydrogen peroxide degradation, it was found that sodium pyruvate was most effective, by a wide margin, in eliminating hydrogen peroxide and its toxic effects. This was confirmed by addition of pyruvate to balanced salt solutions or buffers, and observing the same elimination of hydrogen peroxide. We subsequently found a few earlier reports describing the decarboxylation reaction between hydrogen peroxide and pyruvate, but no kinetic measurements have been published and there seems to be no general appreciation for the very high efficiency of this reaction. The present work presents a preliminary assessment of the importance of pyruvate in the study of hydrogen peroxide and other reactive oxygen species in mammalian cell culture.